.jpg)
(Making data consistent and universally understood isn't as hard as you think...)
Imagine. You’ve got an interview for your dream job at an elite club or National Governing Body. After degree levels of preparation your backside’s shuffling on a seat in front of the interview panel. First question…it’s easy! But question 2 arrives in what your sunshine experiences suggest is Spanish, question 3 pounces on you in Portuguese, question 4 jabs you in Japanese…
The interview is a disaster. Your considerably effort is wasted.
Right now, this situation is playing out for many genuinely data forward sports organisations. Sports data analysis and sports performance analysis is being held back by the bungee cord of data definitions.
Data definitions create the ‘language’ of data. Yet these definitions are often inconsistent between the many (often individually brilliant) systems available to sport.
Examples we’ve seen:
Master Data level (this is top-level data that, if wrong, affects other data attributes)
· Matches classified incorrectly (across event types, age groups). The implications are obvious when it comes to looking at individual level performance data (for example) ‘goals scored’ or ‘runs made.’
· Inconsistent athlete identification. We’ve seen players names vary slightly between systems. Result? All associated stats are wrong.
Measures (such as related to performance)
Inconsistent definitions of goals scored and runs made between systems. For example, a goal in a cup vs. a league vs. different age groups vs. in a starting line-up or as a sub. A run made in different forms of the game.
Derived fields. Derived data fields are created when two or more data items come together to create something new. For example, distance and speed might combine to create both time and acceleration depending upon what you want to know.
Inconsistent derived fields arise because each data solution creates them in different ways. This is, perhaps, the biggest single cause of data definition problems.
Result? Instead of freeing a club or national governing body to do better, data is creating fragmentation and confusion. We call this the ‘WTF!’ data challenge. So do our clients!
So, what are the options?
1. Ignore the issue. Not tenable given that a new data source appears almost weekly as smart tech becomes more prevalent. Many logins, many more definitions…
1. Hold all your data, all systems, at its most granular level. Then create a defined set of data definitions that everyone understands. Use the granular data to make data from all systems ‘fit’ the same definitions.
Our clients are rapidly adopting option 2. Why? They’ve realised that holding their data at it’s most granular level frees them to ‘see’ opportunities that others cannot. They’ve de-coupled their data from front-end user interfaces. In practice, this gives them their own, consistent, and cross system, data asset.
The benefits of this approach are far more than the immediately obvious. A few of these include:
· One set of data definitions. A shared language across their club or sport.
· Derived fields that others don’t have. The ability to ‘see across’ data sets and create derived data fields rather than just take what the vendors sell. For example, the correlation between sleep and how an athlete performs.
· One login for all data sources. There’s no need to remember to engage with multiple systems.
· One ‘standard’ user interface across all data systems. Making if far easier for users to embrace and gain value from data (plus far more likely they will try!). Read more HERE.
· No restrictions on adding new data from great new sources. The data ‘translation’ approach means it’s immediately useful.
· Control over data from an access, compliance and security, perspectives. For example, sharing the right data for the right roles. This makes it far easier for users to see ‘the wood from trees.’
· Control over data from a commercial perspective. This is of huge importance to income. You can read more HERE.
Taking control over your data is faster, cheaper, and easier than you might imagine. Our clients share three things with us:
· The strategy for their organisation
· A list of all their data assets (we can create this for you)
· A list of data users and what they use data for.
Over a few days together we create a ‘data strategy’ that ensure data closely supports the strategy of your organisation. We agree priorities together.
Then, over a just a few months, we solve the data definitions problem. We create your own, owned, and accurate, data asset.
Finally, we create low-code end user apps. These put relevant pieces of the consolidated dataset into the hands of those who can benefit most.
Does it work?
“We’ve been able to bridge the information gap between players, coaches, analysts, parents, at all levels of the club”
Premier League Club, Sports Data Services client.
If you’d like a 30-minute call to explore if your own organisation is ready to take advantage of the data deluge, then give Alan a shout. He’ll set up a call between us.